Unlock instant, AI-driven research and patent intelligence for your innovation.

A Multi-Scale Community Discovery Method in Online Social Networks

A community discovery and social network technology, applied in the field of online social network multi-scale community discovery, can solve high complexity and other problems, achieve small parameter space, reduce the size of the parameter space, and reduce the size of the effect

Active Publication Date: 2020-08-14
XI AN JIAOTONG UNIV
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although methods based on statistical inference models can accurately discover overlapping traditional communities in directed or undirected networks and have a solid theoretical foundation, such methods often have high the complexity of

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Multi-Scale Community Discovery Method in Online Social Networks
  • A Multi-Scale Community Discovery Method in Online Social Networks
  • A Multi-Scale Community Discovery Method in Online Social Networks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0012] The implementation of the present invention will be described in detail below in conjunction with the drawings and examples.

[0013] step 1:

[0014] definition

[0015]

[0016] Indicates the connection mode, reflecting the node v i with v j connection rules. where w ij represents the node v i to v j The edge weights of represents the node v i out degree. then p i =(β i,j ) n×1 reflects the node v i Connection rules with all nodes in the network. v based on connection mode i ,v j The similarity between them is defined as follows:

[0017] PairSim c (v i ,v j )=sig(cos(∠(p i ,p j ))) (2)

[0018] sig( ) represents the sigmoid function, which maps the similarity to the (0,1) range. Further, the similarity between the node v and the node set U based on the connection mode is defined as follows:

[0019]

[0020] Similarity measure based on connection density:

[0021]

[0022] Among them Sim d (u, U) represents the similarity between n...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a multi-scale community discovery method for an online social network to solve the problems that community mining has high calculation complexity in a large-scale online social network and multiple elements are involved in community definition. The method is characterized by fusing user interest and cohesion. The mining method mainly comprises three steps as follows: 1) a network coarsening process; 2) community tag initialization based on a probability model on the coarsened network; 3) estimation of a social tag on the basis of semi-supervised learning on a figure. The method has the advantages that the size of the network is greatly reduced through the coarsening process, the probability model for community detection has a very small parameter space, the probability of searching of the globally optimal solution is greatly increased, and the calculation complexity is reduced; secondly, the community can be defined comprehensively by combining the user interest and the cohesion, and the community mining method based on multiple levels is put forward.

Description

technical field [0001] The invention belongs to the field of computer application technology, relates to data mining, online social network and machine learning, and in particular to a multi-scale community discovery method of online social network. Background technique [0002] There are a large number of potential community structures in online social networks. The nodes in these communities have similar characteristics and play similar roles in the network. The coarse-grained network description based on them can simplify the function and interaction of the entire network. research on its evolution. Therefore, identifying community structures helps us gain a deeper understanding of the nature of networks and the relationship between network structure and its functions. Community detection usually refers to the division of network nodes and edges into different sets. The key to community discovery is community definition, and the definitions of different community struct...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): H04L12/24G06Q50/00
CPCG06Q50/01H04L41/12H04L41/142H04L41/145
Inventor 杜友田苏畅管晓宏王倩
Owner XI AN JIAOTONG UNIV